Skip to contents

Write a faststructure file from a tidy data frame

Source: R/write_faststructure.R
write_faststructure.Rd

Write a faststructure file from a tidy data frame. For biallelic dataset only. Used internally in genometranslator and assigner and might be of interest for users. Prepare quality-controlled biallelic markers and consider missingness, minor-allele frequency, and linkage disequilibrium before export.

Usage

write_faststructure(
  data,
  plink.bed = FALSE,
  pop.levels = NULL,
  filename = NULL,
  ...
)

Arguments

data

A supported genomic file, object, or tidy genomic data frame. Default: data = NULL.

To export in plink BED format. This will write 3 files: .bed, .bim, .fam. For this option to work, the argument data must be a GDS file or object. Default: plink.bed = FALSE.

pop.levels

(optional, string) This refers to the levels in a factor. In this case, the id of the pop. Use this argument to have the pop ordered your way instead of the default alphabetical or numerical order. e.g. pop.levels = c("QUE", "ONT", "ALB") instead of the default pop.levels = c("ALB", "ONT", "QUE"). White spaces in population names are replaced by underscore. Default: pop.levels = NULL.

filename

(optional) The file name prefix for the faststructure file written to the working directory. With default: filename = NULL, the date and time is appended to radiator_faststructure_. Default: filename = NULL.

...

other parameters passed to the function.

Value

A faststructure file is saved to the working directory.

Dependencies

Required package dependencies are declared in DESCRIPTION and installed with genometranslator. Run genometranslator_dependencies() to inspect core packages, optional packages, and external executables.

The standard text output has no additional R-package dependency. Setting plink.bed = TRUE requires the optional Bioconductor package SNPRelate. This writer prepares input files; it does not install or run fastStructure.

Data filtering

This writer does not silently filter markers or individuals. It may validate requirements imposed by the destination format and stop with an informative error when the input is unsuitable. It is the user's responsibility to filter and quality-control the data appropriately for the intended analysis before generating the output. Use radr or another suitable workflow when filtering is required.

References

Raj A, Stephens M, Pritchard JK (2014) fastSTRUCTURE: Variational Inference of Population Structure in Large SNP Datasets. Genetics, 197, 573-589.

Author

Thierry Gosselin thierrygosselin@icloud.com